similar to: sieve-cvs build failing @ 'make'

Hi, dovecot-sieve-1.1.2 + dovecot-1.1.beta13 segfaults with the following sieve filter: --- require ["imapflags"]; if header :contains "subject" ["test"] { addflag "$testflag"; } --- when a message with a subject containing "test" is delivered via dovecot lda. The fault backtrace is: (gdb) run Starting

Backported from CVS HEAD to branch_1_0. You can get it as a patch from http://dovecot.org/patches/1.0/quota-rewrite.diff New features include support for multiple quota roots, so you could have eg. per-user quota and a larger domain quota (although no backend actually supports this yet..). The more important feature is that you can configure specific mailboxes to have a bit more space available

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 on dovecot launch, 'dovecot.log' is created as: -rw------- 1 root testuser 271 2006-09-11 07:58 dovecot.log which causes an apparent perms prob with exim using dovecot's LDA ... what/how do i define/change the LOG's ownership/perms with? it seems to ignore the dovecot.conf umask setting, etc. thx. richard - -- /"\ \

what's the correct format for setting per-user override quotas in a passwd-file? in my "dovecot.conf", i've: protocol imap { mail_plugins = quota imap_quota mail_plugin_dir = /usr/local/dovecot-plugins/imap plugin { quota = maildir:storage=10240 # 10 MB quota limit } in my "imap_user_file", i've: passdb passwd-file { args =

-----BEGIN PGP SIGNED MESSAGE----- Hash: RIPEMD160 bldg dovecot cvs branch_1_0 on osx, i'm seeing a sensitivity to order in protocol statements. e.g., on launch: THIS WORKS FINE: protocol lda { sendmail_path = /usr/local/exim/bin/exim ... plugin { quota = maildir:storage=10240 # 10 MB quota limit } } but, THIS DOES NOT protocol lda { ... plugin { quota =

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hi, building OPENSSH v4.3p2 on OSX 10.4.7, all's fine (config, nake, install, exec) out-of-the-box. so, to be clear, this is NOT a blocker, by any stretch ... that said, however ... if, instead of a don't-monkey-with-it build, prior to ./configure, i: autoreconf -i -f then, a subsequent: ./configure \ --prefix=/usr/local/openssh \

-----BEGIN PGP SIGNED MESSAGE----- Hash: RIPEMD160 hi, i've built dovecot latest cvs on OSX 10.4.7. i'm making a 1st attempt @ trying/failing to get TLS operation up-n-running ... my install's OK: Install prefix ...................... : /usr/local/dovecot File offsets ........................ : 64bit I/O loop method ..................... : poll File change notification method

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hi matthias, reading the dovecot wiki page: http://wiki.dovecot.org/Quota that, apparently, you last edited (created?): Quota (last edited 2006-06-13 15:27:38 by MatthiasAndree) i note: " plugin { # 10 MB quota limit quota = maildir:storage=10240 # 1000 messages quota limit quota = maildir:messages=1000 # 10 MB + 1000

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hi, i build mysql (currently v5.0.22) on OSX 10.4.7 w/ threading enabled. i.e., i build the reentrant libs: libmysqlclient_r.dylib rather than the 'usual': libmysqlclient.dylib currently, to make Dovecot 'play nice' and recognize the reentrant libs, I: perl -pi -e

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hi all, per the exim docs @: http://exim.org/exim-html-4.62/doc/html/spec_html/ch36.html a cyrus_sasl authenticator driveer is 'included' in exim core. given that Dovecot has become a very attractive alternative for an increasing number of folks, and there's a patch against exim 4.43 here:

hi, i'm running dovecot branch_10 cvs on OSX 10.4.7 w/ exim 4.63. i'd like to use Dovecot's SASL as an auth backend to exim, instead of Cyrus-SASL. @ dovecot's sasl page: http://wiki.dovecot.org/Sasl i know there's a patch against exim 4.43: http://www.exim.org/eximwiki/AuthenticatedSmtpUsingDovecot that, purportedly, enables EXIM to use dovecot's SASL as an

Hello, I'm new to dovecot. I just installed it a few days ago on FreeBSD via the ports and got it up and running without too much trouble. Although yesterday I turned on the SSL/TLS features and tried to get secure connections to work. I don't get any error messages when launching dovecot, I saw it successfully create the SSL parameters after first enabling the feature yet I

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hi, i've Dovecot 10branch head built on on OSX 10.4.7. i've also exim 4.63 configured for use with dovecot's AUTH. on local delivery via exim -> dovecot LDA, i see two errors: (1) dovecot_lda transport output: dyld: Library not loaded: /usr/local/ssl/lib/libssl.0.9.8.dylib (2) Child process of dovecot_lda transport (running

hi, i've dovecot cvs 10branch installed on OSX 10.4.7. when i login from my MUA (in this example, Mulberry), functionally all looks OK. BUT, i *do* see one error in the log output, dovecot: Aug 27 20:43:25 Error: IMAP(testuser at mydomain.com): Sending log messages too fast, throttling.. Is this an imap error? or is it 'just' the error of logging too fast? if the latter, what

Hello dovecot users, I have updated the MANAGESIEVE patch to fix the currently known small problems with the protocol implementation. It is designed for rc28, but also compiles cleanly with the current cvs branch_1_0. Change Log V4 ------------- - Added managesieve_implementation_string setting to the managesieve configuration. This can be used to customize the default

i'm setting up for virtual_users w/ maildir. i have in my dovecot.conf: auth default { passdb passwd-file { args = /etc/dovecot/imapd.passwd } userdb static { args = static uid=300 gid=300 home=/var/dovecot/mail mail=maildir:/var/dovecot/mail/%d/%n/Maildir nice=10 } } default_mail_env = maildir:/var/dovecot/mail/%d/%n/Maildir:INDEX=/var/dovecot/indexes/%d/%n in

Hi Duncan, On 11/10/12 15:47, Duncan Sands wrote: > Hi Roel, > >> When going through the internals of the old C back-end, I see that the CBE >> encapsulates arrays into a struct. The source code has the following comment to >> explain this behaviour. >> >> // Arrays are wrapped in structs to allow them to have normal >> // value semantics

Sanjoy, can you clarify this bit about JVM array lengths? Presumably the same pointer can point to two arrays of different lengths during a program's execution. Does this mean that you're relying on invariant.load having function scope? That is, correctness depends on the pointer not being reused for an array of a different length between the first invariant load of that array length in

Hi Roel, > When going through the internals of the old C back-end, I see that the CBE > encapsulates arrays into a struct. The source code has the following comment to > explain this behaviour. > > // Arrays are wrapped in structs to allow them to have normal > // value semantics (avoiding the array "decay"). > > For example, the CBE translates: >

Hello all, When going through the internals of the old C back-end, I see that the CBE encapsulates arrays into a struct. The source code has the following comment to explain this behaviour. // Arrays are wrapped in structs to allow them to have normal // value semantics (avoiding the array "decay"). For example, the CBE translates: @a = common global [10 x i32]